Extract inline vars from the status handler

[mccutchen]

This is getting a bit into micro-optimization territory, but in the course of working on a bit of cleanup I realized there might be something to gain from making the /status endpoint a bit more efficient by initializing all the data it needs once instead of on every request.

Results, from a set of highly unscientific vegeta benchmark runs on my laptop (3x per version):

before

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         599239, 19973.51, 19971.04
Duration      [total, attack, wait]             30.005s, 30.002s, 3.711ms
Latencies     [min, mean, 50, 90, 95, 99, max]  100.397µs, 3.434ms, 2.52ms, 7.487ms, 9.388ms, 14.141ms, 35.997ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:599239  
Error Set:

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         583561, 19451.76, 19449.88
Duration      [total, attack, wait]             30.003s, 30s, 2.905ms
Latencies     [min, mean, 50, 90, 95, 99, max]  90.632µs, 3.596ms, 2.634ms, 7.81ms, 9.657ms, 14.375ms, 71.383ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:583561  
Error Set:

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         583978, 19466.01, 19463.01
Duration      [total, attack, wait]             30.005s, 30s, 4.622ms
Latencies     [min, mean, 50, 90, 95, 99, max]  99.886µs, 3.578ms, 2.633ms, 7.769ms, 9.727ms, 14.462ms, 46.364ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:583978  
Error Set:

after

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         674351, 22478.36, 22476.29
Duration      [total, attack, wait]             30.003s, 30s, 2.765ms
Latencies     [min, mean, 50, 90, 95, 99, max]  88.928µs, 2.487ms, 1.823ms, 5.248ms, 6.888ms, 10.986ms, 57.678ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:674351  
Error Set:

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         664761, 22158.67, 22156.44
Duration      [total, attack, wait]             30.003s, 30s, 3.018ms
Latencies     [min, mean, 50, 90, 95, 99, max]  84.192µs, 2.576ms, 1.904ms, 5.464ms, 7.031ms, 10.955ms, 46.036ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:664761  
Error Set:

$ echo "GET http://localhost:8080/status/200" | vegeta attack -rate 0 -max-workers 100 -duration 30s | vegeta report
Requests      [total, rate, throughput]         668110, 22267.73, 22267.02
Duration      [total, attack, wait]             30.004s, 30.004s, 962.405µs
Latencies     [min, mean, 50, 90, 95, 99, max]  82.35µs, 2.533ms, 1.844ms, 5.384ms, 7.007ms, 11.363ms, 47.509ms
Bytes In      [total, mean]                     0, 0.00
Bytes Out     [total, mean]                     0, 0.00
Success       [ratio]                           100.00%
Status Codes  [code:count]                      200:668110  
Error Set:

That's a decent uptick in request rate and throughput and a decent drop in latency percentiles.